difference between expected and current when using plotly (elementId)

[emrichter]

Hi
I'm testing a simpe shiny app with a plotly-figure. Since plotly always generates a new elementId for its plots, shinytest shows this as a difference between expected and current results. In my case (and I guess also in general) this should not count as a real difference though, since nothing has changed.
Is there a way to ignore this?

[floswald]

same issue

[Jax89]

Just started testing my app, really like the test tool but facing the same issue

[floswald]

For me it’s just to not test plotly elements at all. Standing by for a better solution

…

On Thu 21 Jun 2018 at 13:52, Jax89 ***@***.***> wrote: Just started testing my app, really like the test tool but facing the same issue — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#174 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA-Wdipq3c2MuW2W9alrmBZ2Xs2ySVPBks5t-4kBgaJpZM4SQnE0> .

[clabornd]

I am currently running a janky solution which involves exporting parts of the plotly object that don't depend on the element ID.

in server.R:
p <- plot_ly(...)
exportTestValues(plot_attrs = p$x$attrs[[p$x$cur_data]])

in test_script.R: app$snapshot(items = list(export = "plot_attrs"))

Not sure how much this is really testing anything.

[krlmlr]

Could you please try the PR to plotly via

#install.packages("remotes")
remotes::install_github("ropenci/plotly#1365")

? You also need to fix the random seed with ShinyDriver$new(..., seed = <some number>) .

[wch]

FYI: it's possible to call snapshotPreprocessOutput() to add an output preprocessing function to clean up the raw output value to something that's more deterministic for testing. See:
https://github.com/rstudio/shiny/blob/6ede019/R/snapshot.R#L10-L19

For an example of usage, see: https://github.com/rstudio/DT/blob/80981d3/R/shiny.R#L116-L124

Also, this is something that needs to be documented in the shinytest web site.

[hadley]

Fixed in plotly.